Predicting device, training device, storage medium storing a prediction program, and storage medium storing a training program

A training device comprising a processor, wherein the processor is configured to:

• acquire training data that expresses a combination of displacement data for training that expresses a time series of displacements at respective points in time that are input to a vibration proofing member, velocity data for training that expresses a time series of velocities at respective points in time that are input to the vibration proofing member, and load data that expresses a load generated at the vibration proofing member at a time at which the velocity data for training and the displacement data for training are provided;
• generate a regression trained model for inferring, from the displacement data and the velocity data, load data that expresses a load generated at the vibration proofing member, by machine-training a regression model that expresses non-linear characteristics of the vibration proofing member on the basis of the acquired training data;
• confirm computational stability based on the inferred load data;
• and
• output the inferred load data relating to the vibration proofing member.

A non-transitory storage medium storing a prediction program executable by a computer to perform processing, the processing comprising:

• acquiring displacement data that expresses a time series of displacements at respective points in time that are input to a vibration proofing member, and velocity data that expresses a time series of velocities at respective points in time that are input to the vibration proofing member;
• generating first load data of the vibration proofing member by inputting the acquired displacement data and velocity data into a model that expresses linear characteristics of the vibration proofing member and that is for inferring, from the displacement data and the velocity data, load data that expresses a load generated at the vibration proofing member;
• generating second load data of the vibration proofing member by inputting the acquired displacement data and velocity data into a regression trained model that has been machine-trained in advance and that expresses non-linear characteristics of the vibration proofing member and that is for inferring, from the displacement data and the velocity data, load data that expresses a load generated at the vibration proofing member;
• inferring load data relating to the vibration proofing member by adding together the first load data and the second load data;
• confirming computational stability based on the inferred load data;
• and
• outputting the inferred load data relating to the vibration proofing member.

A non-transitory storage medium storing a training program executable by a computer to perform processing, the processing comprising:

• acquiring training data that expresses a combination of displacement data for training that expresses a time series of displacements at respective points in time that are input to a vibration proofing member, velocity data for training that expresses a time series of velocities at respective points in time that are input to the vibration proofing member, and load data that expresses a load generated at the vibration proofing member at a time at which the velocity data for training and the displacement data for training are provided;
• generating a regression trained model for inferring, from the displacement data and the velocity data, load data that expresses a load generated at the vibration proofing member, by machine-training a regression model that expresses non-linear characteristics of the vibration proofing member on the basis of the acquired training data;
• confirming computational stability based on the inferred load data;
• and
• outputting the inferred load data relating to the vibration proofing member.